RNA polymerase (RNAP), long known as a sophisticated molecular machine in catalysis of templated RNA synthesis, has recently been shown to also function as a molecular isomerization machine to prepare both promoter DNA and RNAP itself for initiation of RNA synthesis. Bending and wrapping of the upstream DNA on the back side of RNAP positions it to interact with downstream mobile elements (DME); this interaction is required to move other DME out the active site cleft and allow RNAP to bend downstream duplex DNA into the cleft for efficient opening. After opening, these same DME assemble in steps into a stabilizing structure that appears to encircle the downstream duplex, creating a series of open complexes that differ greatly in stability and lifetime. The template strand in these open complexes appears to be positioned in the active site; repositioning of the downstream portion of the nontemplate strand in the cleft, possibly facilitated by the ssDNA mimic region 1.1 of ?70, accompanies these conformational changes in the DME. Large conformational changes in DME are reported to occur in the first steps of initiation as well. Our long term goal is to determine the roles of the DME in these large-scale conformational changes in promoter DNA and RNAP that are needed for conversion (classically called isomerization) of the initial promoter recognition (closed) complex to initiation- capable open complexes, and subsequently for transcription initiation and the transition to elongation. Specific aims include: ) Determine the functions of downstream mobile elements (DME) in the early steps of isomerization that wrap upstream DNA, place downstream duplex DNA in the cleft, and open it by characterizing effects of DME deletions on the kinetics of these steps and on the structure of the key intermediate closed complex I1. 2) Determine the functions of the DME in the conversion of the initial unstable open complex (I2) to the stable RPo. 3) Determine the functions of the DME in catalytic steps of initiation, and test the hypothesis that the three open complexes detected at LPR (I2, I3, RPo) are structural and functional analogs of the three classes of open promoter complexes exemplified by rrnB P1, T7A1, and LPR. Experiments are proposed to determine the structural and mechanistic origins of differences in isomerization rate, in properties of open complexes, in initiation rate, and in response to DksA between WT and variant RNAP with deletions in key DME. Methods to be used include footprinting, fluorescence (FRET, quenching assays) and crosslinking studies of transient (unstable) closed and open complexes to characterize them and the rate-determining opening step that relates them. In addition the different open complexes and their putative different states of assembly of the DME will be characterized using solute probes, footprinting, and productive/abortive initiation assays.